Blog & news

Blog — Data Recovery and Cybersecurity

Technical analyses, prevention tips and data recovery news from the experts at SOS Data Recovery, Swiss laboratory since 2006.

Cybersecurity & prevention

In-depth technical analyses

Expert practical advice

Why read our data recovery blog?

The SOS Data Recovery blog covers all topics related to data recovery, cybersecurity and the protection of digital storage media. Our experts share in-depth technical analyses, prevention tips and insights from over 20 years of laboratory experience.

Whether you are an individual looking to protect your photos and documents, or an IT professional managing RAID servers and NAS systems, you will find practical, verified information here. SOS Data Recovery is CyberSafe certified and CyberSafe partner and rated 4.8/5 on Avis Vérifiés (296 reviews).

Our articles : #Backup

Active filter : #Backup All articles

Data loss: the right moves in the first 24 hours

After a data loss, there is no single right move, there is the right move for your problem. This guide helps you recognise which case you are in and act correctly in the first hours. One reflex applies to all: switch the device off quickly (phone, drive, NAS, RAID) and do not tinker with anything before having it analysed. Some situations keep getting worse and require acting fast: water, extinguisher powder and chemicals, which eat away at the hardware even when powered off, as well as ransomware, TRIM on SSD and updates, which run as long as the machine is on. Others, mechanical failure, short circuit, formatting or deleted files, freeze once switched off and can wait. You will also discover the two pitfalls that doom your data: repeatedly restarting a drive and launching a RAID rebuild after an earlier failure.

September 03, 2026 Author Stéphane Chapuis Read More

#Backup #Crisis Management #Data protection #how-to guide #Server

Corrupted Virtual Machine: Recovering a Virtualized Server (VMware, Hyper-V, Proxmox)

A virtual machine is not stored in a partition: it is one large file (a VMDK on VMware, a VHDX on Hyper-V) sitting on a server that is almost always in a RAID array. When the hypervisor crashes, the RAID goes down or this file becomes corrupted, recovery plays out across two layers: first locate the VM file, then rebuild the system and the data it contains. This guide clearly explains the difference between desktop virtualization and a hypervisor (ESXi, Hyper-V, Proxmox), how a virtualized environment fails, and why a snapshot is never a backup. You will also see the mistakes that make everything worse, such as recreating a datastore or launching a RAID rebuild, and the only right reflex in a disaster: shut everything down, reinstall nothing and hand the disks to a lab that works on bit-for-bit copies.

August 25, 2026 Author Stéphane Chapuis Read More

#Backup #RAID #Server

Is the Cloud Really a Backup?

Is the cloud really a backup? No: OneDrive, iCloud, Google Drive, Dropbox and ProtonDrive are synchronisation services, not backup systems. This guide explains why that distinction changes everything. An accidental deletion, a corruption or a ransomware attack spreads instantly to all your copies, including those in the cloud, as in the real case of a company that lost everything after its synchronised OneDrive was encrypted. You will discover everything that can erase your data remotely (closed account, provider bankruptcy, payment incident, a data centre fire like the OVH one), why no provider guarantees the survival of your data, and what truly separates synchronisation from a cold or immutable remote backup. Finally, the one rule that protects for good: keep a physical copy, offline and off-site.

August 07, 2026 Author Stéphane Chapuis Read More

#Backup #Cloud #Data protection #Server

How to Properly Erase Your Data Before Reselling or Discarding Hardware

Deleting a file, emptying the recycle bin or running a quick format only makes the index that points to your data disappear: the data itself stays physically present, and a simple free tool is often enough to restore it. Everything a lab knows how to recover, a curious buyer can recover too. This guide details, medium by medium, how to truly erase your data before reselling or discarding a computer, phone, drive or USB stick: overwriting every sector, factory reset on an encrypted phone, a degausser for tapes, and physical destruction as a last resort. You will see why a discarded medium is never anonymous, how to destroy a chip without piercing the battery, why BitLocker encryption does not remove the need to destroy, and what Switzerland's nLPD and the GDPR require of businesses.

July 29, 2026 Author Stéphane Chapuis Read More

#Backup #Data protection #nLPD

A NAS Is Not a Backup: The Costly Confusion

A NAS is not a backup, and this confusion is one of the most costly in business. Because a NAS fitted with RAID tolerates the failure of a disk, people believe it is safe from everything. Yet RAID protects the hardware, never the data: it will save you neither from a deletion, nor from ransomware, nor from a fire, nor from a corruption that spreads. This guide explains what a NAS really is, the nuance with a SAN, and why redundancy does not replace a copy. You will also understand the trap of synchronisation, which faithfully copies deletions and corruptions, backed by a real-life case where two distant NAS units shared the same problem. Finally, you will see what makes a real backup: a separate copy, disconnected and stored off-site, the only last resort when everything else has failed.

July 02, 2026 Author Stéphane Chapuis Read More

#Backup #RAID #Server

SSD: Why Data Recovery Has Nothing to Do With a Hard Drive

Recovering data from an SSD has almost nothing in common with a hard drive. There are no platters and no read/write heads here, but NAND memory chips driven by a controller, a bit like a USB stick, only far more complex. This difference in nature changes everything: the data is not stored where the system thinks it is, TRIM genuinely erases what you delete within minutes, and hardware encryption can lock everything away if the controller dies. This guide explains how an SSD is built, why recovering one is a trade of its own, and which failures come up most often: faulty controller, corrupted firmware, power surge, cold solder joint. You will also discover the puzzle of invisible BGA solder joints, what a lab really does (tools such as the PC-3000, firmware repair, desoldering and reading the chips), and above all the right moves: stop the device and never run a secure erase.

June 23, 2026 Author Stéphane Chapuis Read More

#Backup #how-to guide #SSD

I Formatted or Deleted by Mistake: Your Real Chances of Recovery (and the Error That Destroys Everything)

Deleting a file or formatting a drive does not physically erase it: on a hard drive, the data stays on the platters as long as no new data has been written over it. Your chances of recovery after an accidental deletion or format are therefore often good, on one imperative condition: stop using the device immediately. This guide explains why deleting is not erasing, how the allocation table (the drive's table of contents) works, and where the real danger lies: overwriting, caused by updates, antivirus or defragmentation that write into the freed space. You will understand the decisive difference between a hard drive and an SSD (where TRIM erases fast), between a quick and a full format, and above all why the recovery software you download can destroy what it claims to save. Finally, how a lab protects your device with a bit-for-bit copy before any intervention.

May 09, 2026 Author Stéphane Chapuis Read More

#Backup #Data protection #how-to guide

RAID or NAS Failure: The Mistakes That Turn a Fault Into Permanent Loss

When a RAID, a NAS or a server fails, it is almost never the initial fault that loses the data, but the repair attempt that follows. This guide explains why the rebuild is the most dangerous moment in a RAID's life: reading every sector of drives of the same age often makes a second drive fail and, on a RAID 5, everything is then lost. You will understand the RAID levels (0, 1, 5, 6, 1+0), why a RAID is never a backup, and the mistakes that kill: deleting and recreating the RAID, multiplying the handling. The right reflex is counterintuitive: switch everything off, rebuild nothing, bring only the drives to a laboratory that works on bit-by-bit copies. Two counterintuitive pieces of advice complete the guide: mix the brands of your drives and steer clear of proprietary hybrid RAIDs.

April 30, 2026 Author Stéphane Chapuis Read More

#Backup #RAID #Server

Your Archives Are Sleeping on Dying Tapes: When and How to Migrate Your Old Backups

A well-stored magnetic tape preserves its data for decades: it is almost never the medium that dies, but everything you need to read it again. This article explains why a perfectly readable LTO, DLT or DAT archive can become inaccessible, and how to migrate before it is too late. You will discover the two major obstacles: finding a compatible and functional drive (the LTO compatibility rule, the headache of SCSI and SAS connectivity), then facing the software wall of the format and the backup catalog, without which you do not know what the tape contains. Through a real case where recovering a Retrospect backup meant digging out an old Mac, you will understand why the right benchmark is around fifteen years, prioritizing regulated data, patents and irreplaceable archives. Finally, the step-by-step course of a professional extraction.

April 21, 2026 Author Stéphane Chapuis Read More

#Backup #best practices #Magnetic tape

Clicking Hard Drive: What the Noise Means and Why You Must Switch It Off

Is your hard drive going click click? This noise is not harmless, but it is not an immediate death sentence either. This guide explains what really happens inside: the read/write heads can no longer read the firmware stored on the platters, the circuit board keeps returning them to the stop of the parking zone, and this back-and-forth produces the clicking. Behind one and the same noise lie four possible causes, from a simple head replacement repairable in a cleanroom to permanent loss when a broken head grinds away the surface. You will understand why every restart to see if it works makes the failure worse, why the freezer, opening the drive or swapping the board must be avoided, and what the only reasonable action is before entrusting the device to a laboratory.

March 25, 2026 Author Stéphane Chapuis Read More

#Backup #Hard Drive #how-to guide

Blog — Data Recovery and Cybersecurity

Why read our data recovery blog?

Our articles : #Backup

Data emergency? We respond immediately.

Connectez-vous sur WhatsApp